Fluvoxamine in prevention of relapse in bulimia nervosa: effects on eating-specific psychopathology.

In a double-blind, placebo-controlled study of 72 patients with bulimia nervosa treated successfully with inpatient psychotherapy, the efficacy of fluvoxamine in maintaining improvement was tested. Fluvoxamine and placebo, respectively, were given over a period of about 15 weeks (2-3 weeks inpatient titration phase, 12 weeks outpatient relapse-prevention [maintenance] phase). The variables assessed concerned bulimic behavior and other aspects of eating disorders, global status, depression, anxieties, obsessive-compulsive behavior, and other aspects of psychopathology. Because the dropout rate was relatively high (N = 27 [33%]) and because it was considerably higher in the fluvoxamine group (19 out of 37 subjects), analyses were performed on the intent-to-treat sample (ideally including all 72 subjects). Results of the completer sample analyses (including only those subjects who finished the study) are briefly presented for comparison. In both the intent-to-treat and the completer analyses, the following scales showed fluvoxamine to have a significant effect in reducing the return of bulimic behavior: (1) self-ratings: Eating Disorder Inventory (EDI)-bulimia, urges to binge in previous week and the number of actual binges in the previous week; (2) expert ratings: Psychiatric Status Rating Scales for Bulimia nervosa, Structured Interview for Anorexia and Bulimia nervosa (SIAB)-"total score," SIAB-subscale "fasting," and SIAB-subscale "vomiting." Two further variables (EDI-total score and SIAB-subscale "bulimia") showed the superior relapse prevention effects of fluvoxamine compared with placebo for the completer sample, while they did not reach significance for group-by-time interactions in the intent-to-treat sample. During a final, short (4-week) off-medication phase, no effect of the discontinuation of medication was observed.